Livestock Detection and Counting in Kenyan Rangelands Using Aerial Imagery and Deep Learning Techniques

Author:

Ocholla Ian A.12ORCID,Pellikka Petri123ORCID,Karanja Faith4ORCID,Vuorinne Ilja12ORCID,Väisänen Tuomas156ORCID,Boitt Mark7ORCID,Heiskanen Janne18ORCID

Affiliation:

1. Department of Geosciences and Geography, University of Helsinki, P.O. Box 64, 00014 Helsinki, Finland

2. Institute for Atmospheric and Earth System Research, University of Helsinki, P.O. Box 4, 00014 Helsinki, Finland

3. Wangari Maathai Institute for Environmental and Peace Studies, University of Nairobi, Nairobi P.O. Box 29053-00625, Kenya

4. Department of Geospatial and Space Technology, University of Nairobi, Nairobi P.O. Box 30197-00100, Kenya

5. Helsinki Institute of Sustainability Science, University of Helsinki, P.O. Box 4, 00014 Helsinki, Finland

6. Helsinki Institute of Urban and Regional Studies, University of Helsinki, P.O. Box 4, 00014 Helsinki, Finland

7. Institute of Geomatics, GIS and Remote Sensing, Dedan Kimathi University of Technology, Private Bag, Dedan Kimathi, Nyeri P.O. Box 10143-10100, Kenya

8. Finnish Meteorological Institute, P.O. Box 503, 00101 Helsinki, Finland

Abstract

Accurate livestock counts are essential for effective pastureland management. High spatial resolution remote sensing, coupled with deep learning, has shown promising results in livestock detection. However, challenges persist, particularly when the targets are small and in a heterogeneous environment, such as those in African rangelands. This study evaluated nine state-of-the-art object detection models, four variants each from YOLOv5 and YOLOv8, and Faster R-CNN, for detecting cattle in 10 cm resolution aerial RGB imagery in Kenya. The experiment involved 1039 images with 9641 labels for training from sites with varying land cover characteristics. The trained models were evaluated on 277 images and 2642 labels in the test dataset, and their performance was compared using Precision, Recall, and Average Precision (AP0.5–0.95). The results indicated that reduced spatial resolution, dense shrub cover, and shadows diminish the model’s ability to distinguish cattle from the background. The YOLOv8m architecture achieved the best AP0.5–0.95 accuracy of 39.6% with Precision and Recall of 91.0% and 83.4%, respectively. Despite its superior performance, YOLOv8m had the highest counting error of −8%. By contrast, YOLOv5m with AP0.5–0.95 of 39.3% attained the most accurate cattle count with RMSE of 1.3 and R2 of 0.98 for variable cattle herd densities. These results highlight that a model with high AP0.5–0.95 detection accuracy may struggle with counting cattle accurately. Nevertheless, these findings suggest the potential to upscale aerial-imagery-trained object detection models to satellite imagery for conducting cattle censuses over large areas. In addition, accurate cattle counts will support sustainable pastureland management by ensuring stock numbers do not exceed the forage available for grazing, thereby mitigating overgrazing.

Funder

European Union DG International Partnerships under DeSIRA (Development of Smart Innovation through Research in Agriculture) programme

University of Helsinki

Publisher

MDPI AG

Reference75 articles.

1. FAO (2018). World Livestock: Transforming the Livestock Sector through the Sustainable Development Goals, Food and Agriculture Organization of the United Nations.

2. Pica-Ciamarra, U., Baker, D., Morgan, N., Zezza, A., Azzarri, C., Ly, C., Nsiima, L., Nouala, S., Okello, P., and Sserugga, J. (2014). Investing in the Livestock Sector: Why Good Numbers Matter. A Sourcebook for Decision Makers on How to Improve Livestock Data, World Bank. Report Number 85732-GLB.

3. Automated Aerial Animal Detection When Spatial Resolution Conditions Are Varied;Brown;Comput. Electron. Agric.,2022

4. The Impact of Drought on Commercial Livestock Farmers in South Africa;Mare;Dev. Pract.,2018

5. FAO (2023). Disasters on Agriculture and Food Security Through Investment in Resilience, Food and Agriculture Organization of the United Nations.

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3